Surgical Innovations

Introduction

Resection of intracranial arteriovenous malformations (iAVMs) has greatly advanced in the past few decades due to innovations in the pre-, intra-, and postoperative settings. With respect to preoperative decision-making, studies have shed light on the natural history of the disease, enhancing patient selection and clarifying indications for surgery. Structural imaging studies such as diffusion tensor imaging and MRI-based volume estimation facilitate more refined assessment of angioarchitecture, structures at risk, and therapeutic targeting. Functional studies, including magnetoencephalography, MR-based perfusion and flow studies, and superselective angiography, combined with Wada testing, allow better characterization of flow dynamics and eloquence of tissue at risk. Improvements in anesthesia, surgical actuators and visualization (including bipolar forceps and surgical microscopes), and neurocritical care have decreased intraoperative and postoperative complications and risks. Incorporation of digital subtraction angiography (DSA) into the operating suite has allowed surgeons to definitively ascertain AVM obliteration intraoperatively, in real time. In addition, operative adjuvants such as indocyanine green and preoperative embolization have also improved the safety of AVM surgery.

Patient Selection: Indications for Surgery and Natural History

Patient selection and subsequent outcomes have undoubtedly been enhanced by our modern understanding of the natural history of AVMs. The prevalence of AVMs in the population is difficult to estimate, as the studies in the literature typically rely on imaging in symptomatic patients rather than a true population cross section. Detection rates in these studies range from 1.12 to 1.42/100,000 person-years, with roughly half of all patients with AVMs suffering intracranial hemorrhages at some point immediately before or after detection. Overall, the risk of AVM-associated hemorrhage is estimated to be between 2% and 4% per year, with increased risk conferred by increased nidus size, flow-related or nidal aneurysms, deep and infratentorial origination, deep venous drainage, cortical venous reflux or outflow obstruction, and prior hemorrhage.

The ARUBA (A Randomised Trial of Unruptured Brain Arteriovenous Malformations) study is the most prominent randomized clinical trial that has directly compared surgical and medical management strategies for iAVMs. The trial enrolled patients who had an iAVM that had been diagnosed on the basis of neurovascular imaging regardless of symptomatic presentation; the patients were randomly assigned to either medical management alone or a combination of medical management and interventional therapy. Some of the patients who were originally assigned to the medical management group underwent interventional treatment and were considered to have crossed over; similarly, patients who were assigned to the intervention group but did not undergo intervention because of an outcome event were considered, for purposes of analysis, to have crossed over to medical management. The breakdown of interventions was as follows: radiotherapy alone (33.0%), embolization alone (31.9%), combined embolization and radiotherapy (16.0%), combined embolization and resection (12.8%), resection alone (5.4%), and all modalities (1.1%). The trial was stopped early due to a larger percentage of individuals in the intervention group experiencing stroke or death (30.7%) compared to the medical group (10.1%). The ARUBA study was criticized for lack of generalizability due to its exclusion of patients with prior hemorrhage, a major risk factor for AVM rerupture, and selection bias (1740 patients screened, 226 enrolled).

Subsequent single-institution data argued that the morbidity risk associated with intervention in ARUBA was overestimated. In addition, surgery, the accepted interventional modality most likely to achieve AVM obliteration, was used in a minority of patients. In a study published in 2018, Link et al. analyzed all of the cases that would have met the ARUBA inclusion criteria and were treated at NewYork-Presbyterian/Weill Cornell Medical Center between 2004 and 2017. The authors identified 86 cases that qualified for the analysis and found a rate of symptomatic stroke or death (the ARUBA primary endpoint) of 8.3% and a rate of long-term impairment of 4.5%. In addition, the nidus was completely obliterated in 92.4% of patients overall and 100% of patients who underwent resection. The American Heart Association and American Stroke Association released a joint statement in 2017 stating that unruptured AVMs could be managed with any combination of microsurgery, embolization, or stereotactic radiosurgery and ruptured AVMs should be managed with a goal of nidus obliteration. To date, there is insufficient high-quality evidence regarding the complications associated with each treatment to issue guidance on the selection of modality for treatment of unruptured AVMs.

Imaging Advances

New imaging modalities have come into practice for preoperative, intraoperative, and postoperative assessment of iAVMs. Currently, DSA with 3D reconstruction is the gold standard for imaging AVMs, allowing full characterization of the lesion with respect to arterial feeders and draining veins. Superselective angiography, employing high-resolution DSA and flexible microcatheters, allows the precise definition of AVM angioarchitecture. When DSA is combined with Wada testing, important functional data can also be obtained. Structural tractography, possibly in combination with functional assessments, allows for determination of AVM involvement with eloquent brain cortex and white matter tracts. If resection of an AVM involving eloquent brain, especially in language or motor areas, is performed, these adjuncts combined with navigation can aid the surgeon during the microdissection. Accurate determination of the nidus volume from MRI is critical for the successful treatment of AVMs, either with microsurgical dissection or stereotactic radiosurgery.

Four-dimensional CT (4D CT) has been shown in a small case series to be almost identical to DSA for determining AVM location, size, and vascular structures. In this study, there was only one reported difference in the identification of smaller arterial branches between the two methods, but 4D CTA was able to distinguish all the main feeding arteries and draining veins. Four-dimensional DSA (4D DSA) allows the contrast bolus to be viewed at any time and from any angle and has been shown to contain information identical to combination of 2D acquisitions and 3D DSA reconstruction. Another study demonstrated that the ability of noninvasive arterial spin labeling (ASL) MRI to identify obliteration of AVMs after stereotactic radiosurgery was similar to that of DSA. This finding may allow patients to avoid the risk of angiography and may enhance patient satisfaction in the surveillance period.

Surgical Advances